Keeping watch on the ocean

See allHide authors and affiliations

Science  23 Feb 2018:
Vol. 359, Issue 6378, pp. 864-865
DOI: 10.1126/science.aar7613

Embedded Image

A commercial fleet fishes for herring in Sitka Sound, Alaska. Real-time information on global fishing activity could guide sustainable ocean management.


Coastal and high-seas fisheries make a crucial contribution to global food security (1), but the vast distances of open waters and limitations of observing systems have precluded a high-resolution global view of fishing activity (2). Monitoring of fisheries has traditionally relied on region- and fleet-specific electronic vessel-monitoring systems, log books, and on-board observers; access to these data is often constrained. Ship positioning data from automatic ship identification systems (AISs), designed to track and monitor specific vessel movements, is providing a surge of information on ocean fisheries. On page 904 of this issue, Kroodsma et al. (3) report tracking industrial fishing from 2011 to 2016 by processing 22 billion AIS messages. They show that more than 55% of the ocean area is fished. Global patterns of fishing are sensitive to culture and political events and are partially insulated from short-term environmental variation.

AIS was developed for collision avoidance. Since 2005, its use has been mandated by the International Maritime Organization for most seagoing vessels of 300 gross tonnage and above and for all passenger ships irrespective of size. Emerging applications of AIS data include fleet and cargo tracking, national fishing fleet monitoring, and maritime security (2, 4). For example, AIS data is yielding information on maritime trade routes and shipping corridors and on trade flows for decision-making (5), enabling assessments of the contribution of ship exhaust emissions to air pollution (6), and allowing evaluation of the exposure of marine mammals and fish to underwater acoustic noise from shipping (7, 8).

Kroodsma et al. used AIS data from 2012 to 2016 to elucidate the spatial and temporal fluctuations of global fishing activities. They generated labeled tracks of more than 70,000 identified fishing vessels by training convolutional neural networks to identify vessel types and fishing activity from AIS; these are tasks that would take a fisheries expert ∼200 years working full time to accomplish (3). Mapping effort over space and time reveals how cultural activities influence fishing activities. The analysis reveals the human face of fishing; for example, latitudinal drops in fishing activity are associated with the Christmas vacation in European and North American fisheries and the Chinese New Year for Chinese vessels. A reduction in activity at weekends is evident in European and North American fisheries effort but less so in Chinese vessels. Political influences are also evident, such as the annual moratoria for Chinese vessels.

Climatic variations can influence the location and abundance of fish stocks and their accessibility to fishing fleets. For example, an abrupt warming in the Western Indian Ocean in 1997 to 1998 changed the tuna distribution; as a result, many fishing fleets shifted to the eastern Indian basin (9). Kroodsma et al. explored the relationships of fishing activity to the elevated sea surface temperatures in the Indian and Pacific Oceans in 2015, the result of a positive Indian Ocean Dipole Mode Index and a large El Niño event. As their analysis shows, longline fishing in the Indian Ocean concentrates between the 16° and 19° isotherms and shifted 70 to 90 km south in July 2015 compared with the preceding and following years. They report that the influence of the large El Niño event was evident in shifts of fishing effort across all fleets in the equatorial Pacific. Kroodsma et al. also found reduced fishing activity in the exclusive economic zones of many island states—information relevant to conservation planning. The availability of oceanographic data has driven the development of dynamic ocean management approaches, which use near real-time data to guide marine spatial management (10). The near real-time tracking of shipping and fishing vessels with AIS can provide critical information on the spatial and temporal distribution of resource users.

Extreme events, such as anomalous warm periods and storms, and rising temperatures in the ocean are affecting the distribution and abundance of marine organisms and redistributing fisheries resources (11). Biodiversity models project a large-scale rearrangement of fish stocks over the coming decades, with decreases in fisheries production in tropical regions (12). High-seas fisheries governance has the potential to reduce the risks from climate change—for example, through international cooperation and the closure of high-seas areas to fishing (11, 12).

Understanding the responses of fishing fleets to cultural, economic, and political drivers, such as regulatory policies, will inform efforts to tackle marine global challenges such as threats to biodiversity from climate change and human use. It will also help to effectively manage the ocean to “conserve and sustainably use the oceans, seas, and marine resources for sustainable development” (United Nations Sustainable Development Goal 14). High-resolution and real-time information on fishing activity data from AIS provides crucial information with which to tackle the multiple objectives of marine resource use and conservation.


Navigate This Article